Humoral immune response to active root resorption with a murine model

Humoral immune response to active root resorption with a murine model Kok Tow Ng, BDS, Gregory J. King, DMD, DMSc, and Frank J. Courts, DDS, PhD Gainesville, Fla. A depression in autoantibody titers to tooth root antigens has been shown to coincide with active root resorption in the dog. Since a murine model would facilitate immunologic studies of root resorption because of the availability of syngeneic and immunodeficient strains, the objectives of this study were to develop a quantitative mouse model for root resorption and to determine if a similar drop in tooth root autoantibodies coincides with active root resorption in this species. Uniform areas of necrosis were created in the periodontal ligaments of lower incisors of 36 male Swiss albino mice by inserting a cryoprobe through a skin incision ( - 8 0 ° C; 5 minutes). Contralateral incisors served as controls. At 0, 3, 5, 7, 10, 14, and 21 days; six mice were killed, and blood and incisors were collected. Relative surface areas of root resorption were quantified with micrographs taken at a standardized position, tilt, and magnification with a scanning electron microscope. Serum autoantibody titers were determined with an enzyme-linked immune sorbent assay with antigen prepared from a 5 mol/L quanidine -HCI-EDTA (pH 5.0) extract of incisor roots that were harvested from syngeneic mice. ANOVA and the paired Student t test were used to compare data at the various time points. No root resorption was evident on control teeth. Localized lesions on treated teeth were found to be of significant size between 7 and 14 days (p < 0.05), but most of these erupted into the mouth by 21 days. Autoantibody titers were reduced by 3 days (p < 0.05), remained depressed until 14 days, and returned to pretreatment levels by 21 days. We conclude that the use of the cryoprobe on mouse incisor periodontal ligament will cause root resorption lesions that are morphologically similar to those in previous reports from traumatically and orthodontically treated teeth and that their sizes can be satisfactorily quantified for use in a biologic assay. Furthermore, the mouse, like the dog, harbors a serum autoantibody to tooth root antigens and this is suppressed during active root resorption. (AM J ORTHOD DENTOFAC ORTHOP 1990;98:456-62.)

T h e mechanism by which permanent teeth resorb in response to trauma is poorly understood. The problem is of considerable significance to orthodontists because a higher prevalence of root resorption has been reported in the orthodontically treated dentition compared to controls, ~-6but there is no direct evidence that root resorption stimulated by orthodontic treatment differs morphologically or histologically from that caused by other forms of mild trauma. 79 Immunologic mechanisms are known to play critical roles in the normal and pathologic resorption of calcified tissues, ~°-~and root resorption is a cardinal sign in tooth transplant rejection, tTas In dogs, titers of serum autoantibodies to tooth root antigens have been shown to

From the Departments of Orthodontics and Pediatric DentistD", University of Florida College of Dentistry, Gainesville, Florida. This work was partially funded by grants from the Greater Miami Academy of Orthodontists, The Southern Society of Orthodontists, and the Florida Orthodontie Association. 811115607

456

drop significantly during active root resorption and only to return to pretreatment levels when the offending teeth were either totally resorbed or surgically removed. 19 Further in vivo studies on the immunology of root resorption would be greatly facilitated by an animal model in which syngeneic and immunodeficient strains are readily available. With respect to these requirements the mouse is ideal. However, further requirements for such a model would also need to be satisfied. These include the following: (1) susceptibility to traumatic root resorption, (2) reproducible methods for the stimulation of uniform trauma to the periodontal ligament., (3) development of lesions that are morphologically similar to those reported in humans and other species, (4) reliable methods for the quantification of root resorption lesions, and (5) exhibition of immunologic changes that are similar to those in other species. Recently tissue changes, including external root resorption, have been described after the application of cold to the mouse incisor root. 2° Although promising, the method was not quantitative, and precise control of

Volume 98 Number 5

Humoral immune response to active root resorption

457

the extent and magnitude of the trauma was not possible. Moreover, whether the mouse has autoantibodies to tooth root antigens and, if so, how they respond to active root resorption is unknown. Therefore the purposes of this study were to adapt the mouse-incisor freezing method to make it quantifiable and to simultaneously measure serum autoantibodies to tooth root antigens. The long-range goal is to investigate several possible immunologic scenarios in traumatic root resorption with this model. These are discussed.

MATERIALS AND METHODS Animals Forty-two male Swiss albino mice (body weight = 25 gm) were anesthetized with intraperitonealinjections of ketamine hydrochloride (Ketaset, Bristol Laboratories, Syracuse, N.Y.) (87 mg/kg) and xylazine (Rompun, BAYVET Division, Miles Laboratories, Inc., Shawnee, Kan.) (13 mg/kg). A midline incision was made below the mandibular symphysis with the overlying skin and fasciae reflected to the left to expose the lateral surface of the left mandible. A cryoprobe (Spembly-Amoils TCC 42 Cry Unit, Spembly Ltd., Andover, England) with a tip diameter of 2.5 mm was then applied to the exposed bone surface, and the tissue temperature was maintained for 5 minutes at - 8 0 ° C. The incisions were closed with 6-0 silk sutures. Six mice at each time point were killed over seven time periods--0, 3, 5, 7, 10, 14, and 21 days. While mice were under appropriate anesthesia, their chest cavities were opened, and the cardiac vessels were snipped with a pair of sharp sterilized scissors. Blood samples and sera were collected for autoantibody determinations. The mice were then decapitated, and the mandibles were dissected and fixed in 10% buffered formalin for 48 hours. This was followed by digestion in 5.25% sodium hypochlorite solution until the incisors could be easily removed from the jaw. The intact incisors were then dehydrated in a graded series of acetone solutions (75%, 95%, and 100%) and stored in an oven at 27 ° C.

Scanning electron microscopy Each incisor was mounted on a stub as shown in Fig. 1. This allowed for viewing of both the medial and lateral surfaces. After mounting, the incisors were coated with gold palladium in a high-vacuum coating unit (Hummer 1, Anatech, Ltd., Alexandria, Va.) for 2.5 minutes and viewed with the scanning electron microscope (SEM) (JEOL USA, Inc., Peabody, Mass.) at 60 ° tilt, 20 kV, and 100 ixA.

Fig. 1. Mandibular incisor mounted for viewing of lateral and medial surfaces in the scanning electron microscope. Ruler in mm.

Micrographs were taken of both medial and lateral surfaces. The areas of root surface that were involved in resorption lesions were quantified manually with the principle of point-counting volumetry. ~8 In this technique, micrographs were placed under a regular point lattice. Points coincident with resorption lacunae were scored. The total points scored, corrected to a standard magnification, gave a measure of the relative extent of the resorption areas. These are referred to as resorption indices.

Serum autoantibody determinations Antigen was prepared from extracted incisor teeth that had been harvested from fresh frozen syngeneic mice, washed in cold phosphate buffered saline (pH 7.4) for 24 hours, homogenized, and then extracted in 5 mol/L quanidine HCI-EDTA (pH 5.0) for 48 hours at 4 ° C. The extracts were cleared by centrifugation, renatured by dialysis against deionized water for 48 hours, and then lyophilized. These preparations were dissolved in carbonate buffer to the desired concentration for assay. Serum autoantibody titers were measured on the samples taken from the 36 experimental animals at sacrifice and on six sham-treated controls. The method is an adaptation of the enzyme-linked immune sorbent assay (ELISA). 2~ In this technique, antigen is bound to the bottom of microtiter wells, and sera are incubated in the wells. If antibodies to the antigen are present in the sera, they will specifically bind to the bottom of the wells. The concentration of antigen-antibody corn-

458

Am. J. Orthod. Dentofac. Orthop. November 1990

Ng, King, and Co'urts

290

Total Resorp~on *

Med Resorption

190

Resorption Index 90

-10

i

i

i

6

10

14

Time (Days) Fig. 2. Areas of root resorption measured on medial and lateral incisor root surfaces and total resorption after exposure to - 80 ° C for 5 minutes with the cryoprobe. Each point represents the mean of six observations.

plexes on the bottom of each well is then determined by the reaction of an antibody to mouse-antibody. This is prepared from another species (e.g., goat). The latter is also congugated with a peroxidase enzyme that allows it to be visualized by a color change reaction, which can be quantified with a spectrophotometer. The ELISA is widely used to measure extremely small concentrations of biologic molecules and can be extremely sensitive, precise, and specific. The wells were first plated with prepared mouse antigen stock and left to incubate overnight at 37 ° C under humid conditions. The next day the wells were washed with PBS (0.15 m o l / L ) - T w e e n 20 buffer with the minimicrowash handwasher (Skatronas, Lier, Norway). Then 350 I.tl of BSA-PBS (1%) was used to block all the wells for 1 hour at 25 ° C. The wells were washed as before, and then 100 i11 of each mouse serum sample (dilution 1 : 100 in 0.05 mol/L Tris buffer) was added to the wells for 2 hours at 37 ° C. The wells were washed again and then 100 I.tl of enzyme-labeled antiglobulin peroxidase conjugated goat anti-mouse immunoglobulin was added and left for I hour at room temperature. The wells were washed a final time, and 200 ttl of a freshly made enzyme substrate solution (ophenylenediamine 4.0 mg and 4 ttl of H202 (30%) in 10 ml of

Fig. 3. Scanning electron micrograph of a mouse incisor at 21 days after freezing. Note that the root resorption lesion has migrated incisaIly and would be above the level of the bone and gingiva.

citrate buffer [pH 4.5]) was added to each well and left to stand for 20 minutes at room temperature. The optical densities of the final solutions were read on a multiscan (Titertek Multiskan MC, Flow Laboratories, McLean, Va.) at 494 nm. The optical densities reflect the amount of substrate hydrolyzed, and this is proportional to the amount of antibody present in each serum sample. Specific immunoglobulin bound per serum sample was determined by subtracting background optical densities. These were quantified from preparations that contained all reagents except serum. As controls for antibody specificity, sera that were previously incubated with tooth root antigen as well as sera to an unrelated bacterial antigen (Bacteroides gingivalis) were also included. The means and SEMs of root resorption areas and autoantibody ratios were calculated for all time points. ANOVA and the Students t test were employed for pairwise comparisons between time points. A significance level of 0.05 was set for all tests. Reliability of the point-counting method was assessed by independently repeating each measurement 10 times and by calculating the mean differences and their SEMs. The reliability of this method is indicated by the 95% confidence limits of the mean differences. RESULTS

Root resorption was observed on both the medial and lateral surfaces of the left incisors. No such surface changes were seen at any time interval on the contralateral control incisors. No changes were seen on the lateral surface at either 3 or 5 days, but these changes began to appear by day

Volume 98 Number 5

H u n m r a l i m n n m e response to active root resorption

15

--:..... 7~:-~ - L

£

....

.

459

. . . . . . .

,, • *.2

Fig. 4. Representative root resorption lesion from a mouse incisor. (medial; day 14); r, resorption; a, mineral deposit; bar, 100 ~tM.

Fig. 5. Representative root resorption lesion from a mouse incisor. (lateral; day 14); r, resorption a, mineral deposit bar, 100 ixM.

7 and increased gradually from day I0 onward. On the medial surfaces, no resorptive changes could be detected on day 3, but minimal changes were present by day 5. The area of resorption gradually increased and peaked at day 10. There was no further increase at day 14 (Fig. 2). Several of the teeth on day 21 were lost to follow-up because the lesions erupted into the mouth (Fig. 3). The changes observed on both surfaces were morphologically similar but differed in the time of onset and in the sizes of the lesions. The medial surfaces first showed resorption lesions and also had larger lesions (Figs. 4 and 5). A common finding on treated teeth was the presence of calcified deposits in and around the areas that were being resorbed (a on Figs. 4 and 5). These appeared to form before the resorption lesions and to cover a larger root surface area at the early time intervals but seemed to disappear at the later times. Autoantibodies to tooth root antigens could be detected in control sera, and these had a sustained and significant (p < 0.05) depression from days 3 to 14, with a return to normal levels by day 21 (Fig. 6).

with the appearance of having been formed by the fusion of smaller ones. Others have also described the morphology of root resorption lesions in human beings 23.24 and rats 7 in response to orthodontic treatment. The topographic features of cementum and dentin resorption reported by these investigators are also similar to those described here. The use of the cryoprobe to induce trauma offers a considerable advantage over the use of liquid nitrogen on cotton swabs 2° or luxation 22 for a quantitative root resorption assay because the extent and magnitude of trauma can be precisely controlled. However, it was not possible to compare the current quantitative findings with those of previous studies because the methods in the latter were descriptive. Quantitative SEM has previously been used with success to document orthodontic root resorption in rodents 7 with results very similar to those of the present investigation. Quantification by SEM is considered to be reliable and valid. Resorption lacunae spread laterally and only deepen significantly at the latest time interval. The depths of the lesions remain very shallow in comparison to the extent of their lateral spread. Although absolute measures of lesion volumes may be more specific, a two-dimensional reproduction of a very shallow object is not expected to introduce systematic errors during quantification with relative measures. The use of a constant tilt for viewing all samples reduces errors caused by foreshortening of the images and increases reliability. In the present investigation, the maximum error of the point-counting method was 0.96 point (the maximum limit of the 95% confidence interval). Most of this error was due to difficulty in defining some lesions, especially during the intermediate stages when

DISCUSSION

The application of cold to the lateral surface of the mandible does induce root resorption as previously reported. 2° Although the earlier studies on traumatic root resorption in the rodent were histologic, and the current one used SEM, the progress of the lesions and their morphologies are similar, 2°'22 first appearing around 3 days and continuing to enlarge for 21 days. The early lesions are characterized by numerous small lacunae, whereas later ones consist of fewer, but larger lacunae

Am. J. Orthod. Dentofac. Orthop. November 1990

460 Ng, King. and Courts O.7-

n,I.i.I

I-

#0.6 O m I-

z 0.5

0.4

.

0

i

2

'

l

4

'

I

6

~

I

8

;

I

'

I

;

|

•

I

10 12 14 16 TIME (Days)

'

I

18

'

I

20

;

I

22

'

i

24

Fig. 6. Autoantibody titers (optical density at 494 nm) from sera of mice stimulated to incisal root resorption with thermal (cold) injury to the lateral periodontal ligament. Each point represents the mean of six observations, and the vertical bars represent the SEMs.

the mineralized deposit was being resorbed. For welldefined lesions, the point-counting error was virtually nil. Determination of the total technique error was not done because it was not practical to remount the samples due to the fragility of the teeth and the possibility of destroying their surface integrity by excessive handling. It is likely that the mounting procedure introduces some error into this method. However, if care is taken to mount the incisors in a constant position for all samples, this should be substantially minimized (Fig. 1). The presence of lesions on the medial surfaces of the treated teeth can only be explained if the extent of the traumatic tissue damage in the periodontal ligament goes around or through the tooth. This idea is supported by the observation that the root resorption lesions appear earlier on the medial surface. Because the lateral surface was adjacent to the cryoprobe tip, the periodontal ligament in that location would have experienced the most severe necrosis. On the other hand, the medial surface would have been more peripheral and would therefore be expected to have less severe tissue damage. The pattern of resorption and removal of the cell-free "hyalinized" zones in orthodontic tooth movement has repeatedly been shown to begin at the periphery and move towards the centerY ~ In this case, the medial surface was peripheral, and the lateral surface was at the center of tissue necrosis. Further support for this interpretation comes from the quantitative SEM studies of cemental cratering that occurs in response to orthodontic treatment. 7 These studies demonstrated that teeth treated with forces of the highest magnitude began developing root resorption lesions at a later time than

those that were treated more moderately. Less freezing time would likely produce more moderate areas of necrosis restricted only to the lateral surface. In this experiment a thin layer of mineralized material on the root surface was seen by day 5 and was subsequently removed by day 21. Using the same method of application of cold to the periodontal ligament, 2° as well as other types of trauma, other investigators have reported similar findings in rodents. Deporter and Brown 25 noted the appearance of wllat they called an "irregular cementum surface" in diseased sites of the marginal periodontium. Michaeli et al. 26 noticed the occurrence of a thick layer of irregular material along the cementum of rat incisors after the application of intrusive loads. It has been suggested that the deposition of this mineralized material could play an important role in the initiation of root resorption, 2° but a mechanism has not yet been defined, and the significance of these deposits remains unclear. This study confirms and extends to the mouse the previous observation that titers of autoantibody to tooth root antigens drop significantly during active root resorption only to return when the resorbing teeth are removed from contact with the circulation. ~9":7 It also demonstrates, for the first time, the exciting possibility that these immunologic changes precede significant development of root resorption lesions rather than merely reflecting their presence. This suggests that these humoral changes may have some predictive value for root resorption. However, it remains to be demonstrated whether the reported changes in serum titers represent a causative antecedent event in root resorption or merely

Vohtme 98 Number 5

reflect the ongoing process. Root resorption experiments that use animals with impairment in the production of these antibodies or with the enhancement of their production would be required for this purpose. There are two alternative interpretations for the titer drop. First, the resorbing tooth could be exposing significant amounts of previously sequestered antigen to the circulation, thereby providing a "sink" into which circulating autoantibody can bind. This could result in a lowering of serum autoantibody titers, if the rate of antibody production were less than the rate at which it binds to the "antigen sink." There is evidence that resorbing roots do, in fact, contain more immune complexes than do nonresorbing roots, z7 Whether or not this specific antibody binds during resorption is currently unknown. It does follow, however, that when the teeth are removed from the circulation, the "antigen sink" ceases to exist, and one could then predict a retum of titers to normal levels. This has now been shown to occur in both the dog 19 and the mouse. A second alternative to explain a significant drop in a tooth root autoantibody that coincides with root resorption would be activation of T-cell suppressors. The linkage between immune cells and bone resorption has become partially understood in the laboratory, but the full spectrum of clinical disorders of this relationship remains to be explored. There is a recent report of malignant osteoporosis that occurs in conjunction with defective immunoregulation that is characterized by decreased immunoglobulin production and marked T-cell suppressor activity. :8 The suggestion was made that hard tissue resorption and antibody reductions may be linked by a common pathway that involves the abnormal functioning of immune cells. Other minor bone abnormalities have been observed in patients with certain types of immunodeficiency, including chondroosseous dysplasia in adenosine deaminase deficiency with severe combined immunodeficiency "-9 and metaphyseal changes in some patients with cartilage hair hypoplasia. 3° Interleukin-1 (IL-I) and tumor necrosis factor alpha (TNF-e0 are both potent stimulators of bone resorption in hematologic malignancy, inflammation, and injury, 3~'32and they are derived from leukocytes. Whether or not either of these are effective stimulators of root resorption remains to be demonstrated. Transforming growth factor beta (TGF-fl) has been shown to be released from bone matrix during resorption and also to have potent immunosuppressive actions on a wide range of cytokine-induced activities. 33 Biochemical similarities between tooth and bone matrices would suggest that a similar mechanism may exist in root resorption. In pathologic conditions in which chronic inflam-

Humoral hnmnne response to active root resorption

461

mation is a feature (e.g., rheumatoid arthritis or periodontal disease), increased bone resorption and bone loss do occur. The possibility that the infiltrating immune cells release cytokines that act locally to stimulate bone resorption is now established. 1°-~6However, similar studies in traumatic root resorption are needed. An altered immune status may affect the balance of cytokine production directly by stimulating those factors that enhance resorption (e.g., ILl, TNF-c0 or indirectly by suppressing factors that normally inhibit the production of the resorption stimulators. For instance, the production of the lymphocyte product, interferon gamma (IFN-',/), which is known to inhibit the production of IL-I and TNF-o~, is significantly depressed in patients with rheumatoid arthritis. 3~-36 In recent years, many osteotropic factors and cytokines have been identified. We are now faced with determining whether any of these play roles in traumatic root resorption. The wide use of syngeneic and immunodeficient strains of mice has proven to be a powerful tool in the elucidation of these mechanisms in bone turnover. Armed with a quantitative murine model for root resorption and with the knowledge that an immunologic alteration occurs concurrently, the study of the link between immunologic mechanisms and root resorption should be facilitated. CONCLUSIONS

1. The use of the cryoprobe can create reproducible and quantifiable root resorption lesions in the mouse incisor. 2. The morphologic and quantitative changes that are associated with these root resorption lesions are similar to those obtained by other methods of inducing trauma, including orthodontic treatment. 3. Reductions in titers of autoantibodies to tooth root antigens during active root resorption are quantitatively similar to those previously reported in the dog. 4. This animal model offers considerable potential for use in studies that are designed to elucidate immunologic mechanisms in traumatic root resorption. We thank Ms. Evelyn L. Clausnitzer (E. M. Facility Specialist) for her assistance with the SEM examifiations, Dr. Elizabeth K. Gesenhues (Orthodontic Resident) for her assistance in the experiment, and Dr. Timothy Wheeler for his helpful advice. We also thank the Department of Comparative Opthalmology for the use of the cryoprobe.

REFERENCES

1. KetchamAH. A preliminaryreport of an investigationof apical root resorption of permanent teeth. Ir,rr J OR'roOD1927;13:97127. 2. BecksH. Root resorption and their relation to pathologicalbone formation. It,rr J OR'roODORALStrRn 1936;22:445.

462

Ng, King, and Courts

3. Hemley S. The incidence of root resorption of vital permanent teeth. J Dent Res 1941;20:133-41. 4. Phillips JR. Apical root resorption under orthodontic therapy. Angle Orthod 1955;25:1-12. 5. Hollender L, Ronn;'rman A, Thilander B. Root resorption marginal bone support and clinical c:rown length in orthodontically treated patients. Eur J Orthod 1980;2:197-205. 6. Rudolph CE. An evaluation of root resorption during orthodontic treatment. J Dent Res 1940;19:367-71. 7. King GJ, Fishlschweiger W. The effect of force magnitude on extractable bone resorptive activity and cemental cratering in orthodontic tooth movement. J Dent res 1982;61:775-9. 8. Kvam E. Scanning electron microscopy of tissue changes on the pressure surface of human premolars following tooth movement. Scand J Dent Res 1972;80:357-68. 9. Andreasen JO. Traumatic injuries of the teeth. Philadelphia: WB Saunders 1981. I0. Coccia PT, Kruret W, Cervenka T, et at. Successful bone marrow transplantation for infantile malignant osteopetrosis. N Engl J Med 1980;302:701-808. 11. Mundy GR, Raisz LG. Cooper RA, Schechter GP, Salmon SE. Evidence for the secretion of an osteoclast stimulating factor in myeloma. N Engl J Med 1974;192:1041-6. 12. Horowitz M, Vignery A, Gerslon RK, Baron R. Thymus-derived lymphocytes and their interactions with macrophages are required for the production of osteoclast-activating factor in the mouse. Proo Natl Acad Sci USA 1984;81:2181-5. 13. Bar-Shavit Z, Teitelbaum SL, Reitsma P, et al. Induction of monocytic differentiation and bone resorption by 1, 25dihydroxy vitamin Dj Proc Natl Acad Sci USA 1983;80:590711. 14. Bhalla AK, Amento ED, Clemens TL, ttolick MF, Krane SM. Specific high-affinity receptors for 1, 25 dihydroxy vitamin Dj in human peripheral blood mononuclear cells: presence in monoeytes and induction in T lymphocyte activation. J Clin Endocrinol Metab 1983;57:1308-10. 15. Gowen M, Wood PD, Ihrie ET, McQuire MKB, Russel RCG. An interleukin-l-like factor stimulates bone resorption in vitro. Nature 1983;306:378-80. 16. Bertolini DR, Neduin GE, Bringurein TS, Smith DD, Mundy GR. Siimulation of bone resorption and inhibition of bone formation in vitro by human tumor necrosis factors. Nature 1986;319:516-18. 17. Reviere GR. Sabet TY, Hoffman RL. Transplantation of tooth buds across a multiple non-H-2 barrier. Transplantation 1971; 12:271-8. 18. Weibel ER, Kestler GS, Scherle WF. Practical sterological methods for morphometric cytology. J Cell Biol 1966;30:23-38. 19. King GJ, Courts FJ. Changes in the titer of tooth root antibodies accompanying traumatic root resorption. In: Davidovitch Z, ed. The biological mechanisms of tooth eruption and root resorption. Birmingham, AL: EBSCO Medic, 1988:365-70. 20. Wesselink PR. Resorption of the mouse incisor after the application of cold to the periodontal attachment apparatus. Calcif Tissue Int 1986;39:11-21. 21. Engvall PA, Perlmarm P. Enzyme-linked immunosorbent assay

Am. 3.

Orthod. Dentofac. Orthop. November 1990

(ELISA) quantitative assay of immunoglobulin G. Immunochemistry 1971;8:871-4. 22. Birkedal-Hansen H. External root resorption caused by luxation of rat molars. Seand J Dent Res 1973;81:47-61. 23. Barber AF0 Sims MR. Rapid maxillary expansion and external root resorption in man: a scanning electron microscope study. AM J OR'moo 1981;79:630-51. 24. Reitan K. Initial tissue behavior during apical root resorption. Angle Orthod 1974;44:68-82. 25. Deporter DA, Brown DY. Fine structural observations on the mechanism of loss of attachment during experimental periodontal disease in the rat. J Periodont Res 1980;15:304-13. 26. Michaeli Y, et al. Recovery of the dental and periodontal tissues of the rat incisor following application of continuous intrusive loads: a long-term study. Ara J OR'roOD 1985;87:135-43. 27. King GJ, Courts FJ. Humoral immune responses to active root resorption. In: Norton LA, Burstone CJ, eds. Biological basis of tooth movement. Boca Raton, FL: C.R.C. Press, 1988:27586. 28. Rubin KR, Ballow M, Baron R, Greenstein RM, Raisz LG, Rowe DW. Malignant osteoporosis and defective immunoregulation. J Bone Miner Res 1988;3:509-16. 29. Cederbaum SD, Kartila I, Rimom DL, Stiehm ER. The chondroosseous dysplasia of adenosine deaminase deficiency with severe combined immunodeficiency. J Pediatr 1976;89:737-42. 30. Horowitz SD, Hong R. The pathogenesis and treatment of immunodeficiency. Basel: S. Karger, 1977:91. 31. Gowen M, Mundy GR. Actions of recombinant interleukin 1, interleukin 2 and interferon-gamma on bone resorption in vitro. J lmmunol 1986;136:2478-82. 32. Bertolini DR, Nedwin GE, Bringman TS, Smith DD, Mundy GR. Stimulation of bone resorption and inhibition of bone formation in vitro by human tumor necrosis factors. Nature 1986;319:516-18. 33. Pfeilschifter J, Mundy GR. Modulation of Type B transforming growth factor activity in bone cultures by osteotropic hormones. Proc Natl Acad Sci USA 1987;84:2024-8. 34. Stolzenbury T, Binz H, Fontana A, Felder M, Wagenhaeuser FJ. Impaired mitogen-induced interferon-gamma production in rheumatoid arthritis and related diseases. Seand J Immunol 1988;27:73-81. 35. Seitz M, Napierski 1, Augustin R, Hunstein W, Kirchner H. Reduced production of interferon alpha and interferon gamma in leukocyte cultures from patients with active rheumatoid arthritis. Scand J Rheumatol 1987;16:257-62. 36. Ridley MG, Panayi GS, Nicholas NS, Murphy J. Mechanisms of macrophage activation in rheumatoid arthritis: the role of gamma interferon. Clin Exp Immunol 1986;63:587-93. Reprint requests to: Dr. Gregory J. King Department of Orthodontics University of Florida College of Dentistry J. Hillis Miller Health Center Box J 4~4 Gainesville, FL 32610